Upload New File

Yolo-Obj-Detection/yolo_opencv_videoCopy.py

+#############################################
+# Object detection - YOLO - OpenCV
+# Author : Arun Ponnusamy   (July 16, 2018)
+# Website : http://www.arunponnusamy.com
+############################################
+
+
+import argparse
+import csv
+import cv2
+import numpy as np
+import pandas as pd 
+
+
+#Ref for mov to mp4: http://www.legendu.net/en/blog/python-opencv-python/
+
+VIDEO_STREAM = 'Input_Videos\eyediap\A_1_A_FT_M.mov'
+VIDEO_STREAM_OUT = 'Processed_Videos\Datacollection\output_1_A_FT_M2.mov'
+fourcc = cv2.VideoWriter_fourcc(*"XVID")
+
+cap = cv2.VideoCapture(VIDEO_STREAM)
+ret, frame = cap.read()
+# frame = cv2.resize(frame, (640,480), interpolation=cv2.INTER_AREA)
+
+
+writer = cv2.VideoWriter(VIDEO_STREAM_OUT, fourcc, 30, (frame.shape[1],frame.shape[0]), True)
+
+
+ap = argparse.ArgumentParser()
+ap.add_argument('-i', '--image', required=True,
+                help = 'path to input image')
+ap.add_argument('-c', '--config', required=True,
+                help = 'path to yolo config file')
+ap.add_argument('-w', '--weights', required=True,
+                help = 'path to yolo pre-trained weights')
+ap.add_argument('-cl', '--classes', required=True,
+                help = 'path to text file containing class names')
+args = ap.parse_args()
+
+
+def get_output_layers(net):
+    
+    layer_names = net.getLayerNames()
+    
+    #output_layers = [layer_names[i[0] - 1] for i in net.getUnconnectedOutLayers()]
+    output_layers = [layer_names[i - 1] for i in net.getUnconnectedOutLayers()]
+
+    return output_layers
+
+
+def draw_prediction(img, class_id, confidence, x, y, x_plus_w, y_plus_h):
+    print('x - ',x, 'y- ', y, 'x+w = ', x_plus_w, 'y+h = ', y_plus_h)
+    # print('x1y1 = ',x1_y1, 'x2y2 = ', x2_y2)
+
+    # label = str(classes[class_id])
+    label = 'ball'
+
+    # color = COLORS[class_id]
+    color = 255,0,0
+
+    img = cv2.rectangle(img, (x,y), (x_plus_w,y_plus_h), color, 2)
+    img = cv2.putText(img, label, (x-10,y-10), cv2.FONT_HERSHEY_SIMPLEX, 0.5, color, 2)
+
+    return img
+
+# image = cv2.imread(args.image)
+
+Width = frame.shape[1]
+Height = frame.shape[0]
+scale = 0.00392
+
+print(frame.shape)
+classes = None
+
+frame_count = 0
+df_center_x,df_center_y, df_width, df_height, df_frame_num, df_x1, df_y1, df_x2, df_y2 = [],[],[],[],[],[],[],[],[]
+
+net = cv2.dnn.readNet(args.weights, args.config)
+# COLORS = np.random.uniform(0, 255, size=(len(classes), 3))
+
+while True :
+  ret, image = cap.read()
+  
+
+  # print(ret)
+  if ret:
+    # if frame_count == 10:
+    #     break
+    image = cv2.resize(image, (640,360), interpolation=cv2.INTER_AREA)
+
+    frame_count += 1
+    print(frame_count)
+    center_x, center_y, x, y,w,h = 0,0,0,0,0,0
+
+    with open(args.classes, 'r') as f:
+        classes = [line.strip() for line in f.readlines()]
+
+    blob = cv2.dnn.blobFromImage(image, scale, (416,416), (0,0,0), True, crop=False)
+
+    net.setInput(blob)
+
+    outs = net.forward(get_output_layers(net))
+
+    class_ids = []
+    confidences = []
+    boxes = []
+    conf_threshold = 0.5
+    nms_threshold = 0.4
+
+    for out in outs:
+        for detection in out:
+            scores = detection[5:]
+            class_id = np.argmax(scores)
+            confidence = scores[class_id]
+            if confidence > 0.5 and (class_id == 32 or class_id == 29 or class_id == 47 or class_id == 41 ):
+            # sports ball, frisbee, apple 
+            # and class_id == 24:
+                center_x = int(detection[0] * Width)
+                center_y = int(detection[1] * Height)
+                print(class_id, center_x, center_y)
+                w = int(detection[2] * Width)
+                h = int(detection[3] * Height)
+                x = center_x - w / 2
+                y = center_y - h / 2
+                class_ids.append(class_id)
+                confidences.append(float(confidence))
+                boxes.append([x, y, w, h])
+               
+
+    df_frame_num.append(frame_count)
+    df_width.append(w)
+    df_height.append(h)
+    df_center_x.append(center_x)
+    df_center_y.append(center_y)
+
+
+    x1 = int(x)
+    y1 = int(y)
+    x2 = int(x+w)
+    y2 = int(y+h)
+    x1_y1 = (x1, y1)
+    x2_y2 = (x2, y2)    
+
+    df_x1.append(x1)
+    df_y1.append(y1)
+    df_x2.append(x2)
+    df_y2.append(y2)
+
+    print(x1, ',', y1, '-----',x2, ',', y2 )
+    print("Center x:", center_x, "  center y:", center_y)
+
+
+    indices = cv2.dnn.NMSBoxes(boxes, confidences, conf_threshold, nms_threshold)
+
+    for i in indices:
+        # i = i[0]
+        box = boxes[i]
+        x = box[0]
+        y = box[1]
+        w = box[2]
+        h = box[3]
+        img = draw_prediction(image, class_ids[i], confidences[i], round(x), round(y), round(x+w), round(y+h))
+
+
+    #cv2.imshow("object detection", img)
+    
+    cv2.waitKey(10)
+
+    writer.write(image)
+  else :
+    break
+
+
+
+df_dict = {'Frame_number': df_frame_num, 'Detector_center(x)': df_center_x, 'Detector_center(y)': df_center_y,'detector_bb_height': df_height, 'detector_bb_width':df_width, 'Detector_x1':df_x1, 'Detector_y1':df_y1, 'Detector_x2':df_x2, 'Detector_y2':df_y2}
+df = pd.DataFrame(df_dict)
+print(df.head)
+df.to_csv("csv/csv_eyediap/detector_output_1_A_FT_MTest.csv", index=False)
+
+# VIDEO_STREAM.release()
+writer.release()
+cv2.destroyAllWindows()
+